ORIGINAL RESEARCH published: 18 February 2020 doi: 10.3389/fspor.2020.00007

Wide As a Potential Knee Injury Risk Factor in Alpine

Martin Zorko 1, Bojan Nemec 2, Zlatko Matjaciˇ c´ 3, Andrej Olenšek 3, Katja Tomazin 4 and Matej Supej 4*

1 Clinical Institute of Occupational, Traffic and Sports Medicine, University Medical Centre Ljubljana, Ljubljana, Slovenia, 2 Department for Automatics, Biocybernetics and Robotics, Jožef Štefan Institute, Ljubljana, Slovenia, 3 Research and Development Unit, University Rehabilitation Institute, Ljubljana, Slovenia, 4 Faculty of Sport, University of Ljubljana, Ljubljana, Slovenia

Alpine skis with wider waist widths have recently become more popular. With such skis, the contact point of the ground reaction force during turns is displaced more medially from beneath the sole of the outer ski, which may present an increased risk of injury. The aim of this study was to investigate knee joint kinetics, kinematics, and lower limb muscle activation as a function of changes of the ski waist width in a laboratory setting. A custom skiing simulator was constructed to enable simulation of different ski waist widths in a quasi-static ski turn position. An optical system was used for capturing knee joint kinematics of the outer leg, whereas a force plate was used to determine the ground reaction force vector. The combination of both systems enabled values for Edited by: external torques acting on the knee joint to be calculated, whereas electromyographic Josef Kroell, measurements enabled an analysis of knee flexor muscle activation. With respect to the University of Salzburg, Austria outer ski, the knee joint external torques were independent of ski waist width, whereas Reviewed by: Peter A. Federolf, knee joint external rotation and biceps femoris activation increased significantly with the University of Innsbruck, Austria increase of the ski waist width. Skier muscle and kinematics adaptation most probably Matthias Gilgien, Norwegian School of Sport took place to diminish the external knee joint torque changes when the waist width of the Sciences, Norway ski was increased. The laboratory results suggest that using skis with large waist widths *Correspondence: on hard, frozen surfaces may change the load of knee joint surfaces. However, future Matej Supej research is needed to clarify if this may result in the increased risk of knee injury. [email protected]

Keywords: 3D kinematics, simulation, biomechanics, electromyography, injury prevention, kinetics, ski geometry Specialty section: This article was submitted to Sports Science, Technology and INTRODUCTION Engineering, a section of the journal Alpine skiing is a demanding sensory motor task with respect to maintaining balance and Frontiers in Sports and Active Living counteracting high external forces acting on the human body. It is well-documented that Received: 09 April 2019 recreational alpine skiing is associated with high injury rates ranging from 2.4 to 7.0 injuries per Accepted: 14 January 2020 1,000 activity days (Hebert-Losier and Holmberg, 2013). Such injury rates are several times higher Published: 18 February 2020 in competitive skiing compared with recreational skiing (Haaland et al., 2016). In addition to Citation: frequent acute injuries, overuse injuries are also common (Hildebrandt and Raschner, 2013; Sporri Zorko M, Nemec B, Matjaciˇ c´ Z, et al., 2015, 2016; Supej et al., 2017). The knee joint is the most frequently injured body part in both Olenšek A, Tomazin K and Supej M (2020) Wide Skis As a Potential Knee recreational and competitive skiing (Brucker et al., 2014; Haaland et al., 2016). More specifically, Injury Risk Factor in Alpine Skiing. the outer knee in the ski turn has been reported to be more prone to acute injury compared with Front. Sports Act. Living 2:7. the inner one (Urabe et al., 2002). The relationships between sudden unexpected events and/or loss doi: 10.3389/fspor.2020.00007 of balance accompanied by undesirable lower limb/knee movements while skiing and acute knee

Frontiers in Sports and Active Living | www.frontiersin.org 1 February 2020 | Volume 2 | Article 7 Zorko et al. Wide Skis and Knee Injury

FIGURE 1 | In a ski turn on hard frozen snow, the contact point of the ground reaction force (GRF) of the outside ski is shifted more medially with a wide ski (right: dw) than with a narrow one (left: dn).

injuries have been well-established (Bere et al., 2011; Shea et al., 2014). However, the influence of ski equipment on chronic degenerative knee joint conditions is not yet clear. Human gait analysis studies have proven that even small FIGURE 2 | Skiing simulator placed on the force plate. The platform of the changes in knee joint alignment in the frontal plane, such as simulator was able to freely rotate around the sagittal axis. Two extremes of varus or valgus deformity, can substantially change the loading of the four possible ski width positions are represented (W0 represents the different joint compartments and hasten degenerative processes position of the sagittal axis beneath the middle of the sole, and W120 of the joint (Sharma et al., 2001; Levine and Bosco, 2007). In represents the simulation of waist width of 120 mm). addition, it has been shown that the human body tends to adapt to changes in body segment malalignment and asymmetrical loading of knee joint compartments (Mündermann et al., 2005). Nowadays, skis with very wide waist width on frozen compact been demonstrated (Zorko et al., 2015), the influence of ski waist snow are commonly used, although they were initially designed width on knee loading remains uninvestigated. for “soft-snow” conditions, that is, powder and/or off- Therefore, the aim of the present study was to examine the skiing. A previous study demonstrated that for ski turns on a knee joint kinetics, kinematics, and lower limb muscle activation hard snow base, an increase in ski waist width triggers adaptive with respect to differences in ski waist width. It was hypothesized changes of the knee kinematics in frontal and transversal planes that the wider waist width of the ski has a significant impact (Zorko et al., 2015). This field study showed that knee external on knee position, external knee torque, and muscle activation. rotation increased with wider skis and that the knee joint was in The experiment was conducted in a laboratory setting with valgus/abducted position throughout the ski turn, but, somewhat the purpose of establishing more controlled conditions where controversially, the valgus was smaller with wider skis. The vibrations (Klous et al., 2010) and other influences of uneven main limitation of the study was that the degree of knee joint terrain (Federolf et al., 2009; Supej, 2013) are excluded. flexion while turning was uncontrolled and differed substantially between volunteers. As knee joint accessory movements (rotation and valgus/varus) are coupled with the primary movement of MATERIALS AND METHODS knee in the sagittal plane (flexion/extension) to some degree, the influence of ski waist width on the knee was somewhat unclear Measurement System and uncertain. For the purpose of the study, a specially designed computer- With respect to the outer and more loaded ski involved in controlled test device was built, enabling simulation of different the turn (Vaverka and Vodickova, 2010), the point of application ski waist widths for an outer leg in the quasi-static position of the ground reaction force (GRF) moves from beneath the of a ski turn (Figure 2). The test device comprised a movable foot in the medial direction toward the ski edge (Figure 1). plate equipped with ski bindings that allowed rotation around The greater the ski waist width, the greater the shift of GRF the sagittal axis in order to mimic the behavior of the outer will be, assuming hard snow conditions where the ski does not ski. The inclination angle was measured with an inbuilt absolute sink into the snow (Federolf et al., 2010). Without changing magnetic encoder (RLS Renishaw RM22, Ljubljana, Slovenia). knee kinematics, it can also be assumed that the alignment of The test device was placed on a force plate (Kistler 9253A11, GRF with respect to the center of the joint changes with the Kistler Group, Winterthur, Switzerland) measuring GRF vector different ski waist width and accordingly also its torque. This and its point of application. Radial force (Fr) was simulated hypothetical shifting of the GRF with respect to the center of the and measured using a special harness attached to a load joint in relation to the separate knee compartment may represent cell (S9M/2 kN, HBM GmbH, Darmstadt, Germany). Eleven a critical overload, because the magnitude of GRF reaches 1.5 to2 reflective markers were placed in the following way: six on the times the body weight, even in recreational skiing (Scheiber et al., outer leg, two on the ski boot, and three on the movable plate of 2012), and up to four times the body weight in competitive skiing the simulator (Figure 3); and they were captured by a system of (Supej et al., 2004). Although the relationship between knee joint three calibrated infrared cameras (Optitrack V120:Trio, Natural kinematics and the width of the skis during ski turns has already Point, USA).

Frontiers in Sports and Active Living | www.frontiersin.org 2 February 2020 | Volume 2 | Article 7 Zorko et al. Wide Skis and Knee Injury

markers and EMG electrodes were applied to the subjects (as described in Measurement System) prior to measurement. Initial measurements for surface EMG of the selected muscles during maximal isometric knee extension [maximal voluntary contraction (MVC)] were taken. For isometric measurements, participants sat in an isometric torque measuring device (own design) equipped with a strain gauge force sensor (MES, Maribor, Slovenia) set at the constant lever arm to the rotational axis of the knee. Subjects were seated in an isometric knee torque ◦ ◦ ◦ measuring device with left knee at 60 and 40 (0 full extension), respectively, for knee extensor and flexor MVC assessment, and ◦ hips were at 110 for each. The lower leg was attached to the isometric frame above the ankle joint (i.e., lateral malleolus). The rotational axis of the frame was visually aligned to the rotational axis of the knee (i.e., lateral femoral epicondyle) for each subject prior to measuring. Subjects performed a specific warm-up protocol prior to data collection, consisting of two submaximal FIGURE 3 | Participant bounded to ski turn simulator. The moving platform, isometric knee extensions for 5s at 40, 60, and 80% of MVC optical markers, and the lateral supporting strap are seen. The frontal plane force. Each submaximal isometric knee extension was followed was calculated using three markers placed on the moving platform that by 10-s rest. This was followed by two maximal isometric knee defined two unit vectors. The first vector a was perpendicular, and the second extensions (∼5 s) separated by 1 min of rest. During maximal vector b was parallel to the simulator’s axis of rotation. One coordinate of the frontal plane was defined by vector b, and the other by the vector trials, subjects were instructed to increase torque production over perpendicular to a and b (cross product of vectors a × b). a 2-s period and maintain it for 3 s. The protocol for maximal isometric knee flexion followed exactly the same procedures as for extensors. Following knee extensor MVC assessment, subjects received 3min of rest to avoid fatigue. The highest values from During quasi-static skiing positions on the simulator, and the first two maximal knee extensor and flexor torques for each during maximum isometric contractions that were used for participant were retained for an analysis. Maximal torque was subsequent normalization, surface electromyography (EMG) defined as the maximum value recorded over a period of 0.5 s, recordings were taken from the biceps femoris (BF) and after torque had reached a plateau. The corresponding root mean semitendinosus (ST) muscles by a pair (2.5-cm inter-electrode square (RMS) EMG for the selected muscles was quantified distance) of bipolar self-adhesive electrodes (Ag-AgCl, type during the same 0.5-s period. Average RMS values from the two H124SG, Kendall). The electrodes were placed along the maximal contractions were retained for normalization of surface presumed direction of the underlying muscle fibers according EMG obtained during skiing simulations. to recommendations by SENIAM (Hermens et al., 2000) with Participants then bound their left (outside turn) ski boot the reference electrode placed over the patella. The electrodes into the simulator and were required to maintain the skiing were placed over the muscle belly. Electrodes for BF were posture by inclining the left “ski” to compensate for the simulated placed at 50% on the line between the ischial tuberosity and radial force (Figure 3). The inclination for the center of mass the lateral epicondyle of the tibia. For the ST muscle, electrodes in the frontal plane was calculated from the force platform and were placed at 50% on the line between the ischial tuberosity the load cell measurements and was adjusted to an angle of ◦ and the medial epicondyle of the tibia. Additionally, manual 25 . For subjects standing on one leg, this inclination produced muscle testing was performed to ensure appropriate electrode magnitudes of forces on the outside leg that were comparable placement. Before the electrodes were placed, the skin was with those expected in recreational skiing and were slightly shaved, cleaned, and abraded in order to optimize electrode–skin less than those in competitive skiing (Vaverka and Vodickova, contact and gain low impedance (<5 k) between electrodes. 2010; Scheiber et al., 2012; Gilgien et al., 2014). It would EMG data were recorded with a PowerLab system (16/30- have been interesting to simulate greater skier inclination, but ML880/P, ADInstruments, Bella Vista, Australia) at 2,000-Hz pilot measurements with the participants exposed considerable sampling frequency, amplified by octal bio-amplifiers (ML138, difficulty in maintaining the balance and position of the outer leg ADInstruments) with a bandwidth frequency ranging from 3 to on the widest skis. Real-time visual feedback for the magnitude 1,000 Hz (input impedance = 200 M, common-mode rejection of the inclination of the ski into the turn and for the magnitude ratio = 85 dB, gain = 1,000), and analyzed using LabChart6 of the knee flexion angle was actively used to maintain ski ◦ ◦ software (ADInstruments). inclination at 25 , as well as for knee flexion at 40 in the first and ◦ 60 in the second part of the experiment. After each participant Participants and Measurement Protocol achieved the appropriate posture, a preprogrammed protocol of Fourteen healthy participants (8 males and 6 females; age: 27.9 ski width adjustments was activated. The simulated waist widths ∗ ± 5.4 years; height: 174 ± 9cm; weight: 73.9 ± 9.8kg) with of the ski were 0 (W0), 60 (W60), 90 (W90), and 120 (W120) ∗ intermediate skiing skills participated in the study. Reflective mm; 0 (W0) represented the initial position where the axis of

Frontiers in Sports and Active Living | www.frontiersin.org 3 February 2020 | Volume 2 | Article 7 Zorko et al. Wide Skis and Knee Injury rotation was aligned with the middle of the ski and ski boot in joint, based on the optical marker positions and the standard the mediolateral direction. This position served as the starting anatomical knee model, it was possible to calculate the moment point of the experiment, creating practically zero torque on the arm (r) of the GRF acting on the knee joint. Finally, the vector ski around its longitudinal axis, which does not imitate any product multiplying r and the magnitude of the GRF resulted realistic skiing situation. A motorized tray moved the plate with in the torque value M. In addition, from the ratio between the ski boot in the aforementioned ski width positions every 8 s the vertical (Fv) and horizontal (Fh) components of GRF, the in random order. After four adjustments of the ski waist width, inclination of the skier’s center of mass was calculated as φ = there was a 2-min resting period. The procedure was repeated arctan(Fh/Fv). ◦ ◦ five times for 40 and 60 of knee flexion angles. The calibration All measured EMG signals were filtered with a band-pass procedure for standing upright at 0-mm ski width position was filter (10–500 Hz). An RMS envelope for the EMG signal was made prior to each skiing simulation. calculated using a 300-ms running window. The normalization of the EMG signal was made using the signal at MVC as a Data Processing reference (detailed description presented above). In this way, Data from the final 2 s of the 8-s measurement periods of the quantity of electrical activity for the selected muscles during each trial and waist width were used for further analysis. For the skiing simulation in comparison with the activity at MVC kinematic system measurement, standard Euler’s angles in three was obtained. anatomical planes (Grood and Suntay, 1983) for the simulated outside (left) leg knee joint were acquired in accordance Statistical Analysis with International Society of Biomechanics recommendations. The data are represented as mean and standard deviation Relative changes of the knee joint angles with different ski (mean ± SD). Normality of parameter distribution was waist widths were calculated by subtracting the data acquired at tested using the Kolmogorov–Smirnov test. One-way repeated- calibrating upright postures. measures ANOVA was conducted for normally distributed data. The outside torque (M) acting on the knee joint was Bonferroni corrected post hoc tests were made to determine the calculated in the frontal plane, whose definition was based on between mean differences in the cases where a significant main the orientation of the platform (Figure 3) in each instance effect was seen. In cases where the criteria for repeated-measures of time. With the use of GRF and the center of the knee ANOVA were not met, the Friedman non-parametric test was

FIGURE 4 | External knee rotation (mean ± SD) with knee flexion 40◦ (A) and 60◦ (B). A statistical significant increment of the external rotation was observed with each increase in the ski waist width in both knee flexion settings. Knee abduction/valgus (mean ± SD) with knee flexion 40◦ (C) and 60◦ (D). The statistically significant difference for abduction is marked with “*”.

Frontiers in Sports and Active Living | www.frontiersin.org 4 February 2020 | Volume 2 | Article 7 Zorko et al. Wide Skis and Knee Injury

FIGURE 5 | The magnitude of the ground reaction force (GRF; mean ± SD) FIGURE 6 | The inclination of the ground reaction force (GRF; mean ± SD) with different waist widths for both knee flexion settings. There were no with different waist widths for both knee flexion settings. Throughout the statistical differences in GRF between different ski waist widths. experiment, there were statistically insignificant changes from the preset GRF angle (25◦ from the vertical line). used, and paired comparisons were made using Wilcoxon signed- rank tests. In all cases, a P-value of 0.05 was accepted as the level Electromyography of significance. ◦ BF RMS at 40 knee flexion increased throughout the waist width spectrum (Figure 8, Table 1). Likewise, BF RMS at knee flexion RESULTS 60 increased significantly from 60- to 120-mm waist width. ST Knee Joint Kinematics RMS remained more or less the same for all waist widths and knee flexion values. In all experimental conditions, the knee joint was in external rotation (Figure 4). The Friedman non-parametric test and all pairwise comparisons revealed significant differences between DISCUSSION different waist widths for both of the knee flexion experimental conditions, 40◦ and 60◦. The increment of external rotation was The main findings of the study were as follows: (1) the width of the highest for waist widths between W0 and W60 with further the ski has no influence on the external torque acting on knee increments of external rotation following an almost linear pattern joint in frontal plane; (2) activation of BF muscle increased by the for larger waist width (60 < 90 < 120 mm). waistwidthoftheskiintheskiturn;and(3)kneejointkinematics The knee joint was positioned in abducted (valgus) position in the transversal plane was influenced by the waist width of the throughout the experiment (Figure 4). Generally, there were no ski during a turn. differences between the magnitude of abduction with different waist widths. The only exception was a statistically significant Knee Joint Kinematics lower abduction with W90 (4.54 ± 1.96◦) compared with W0 The knee joint of the outer leg in the simulated ski turn was in (6.84 ± 3.03◦, P < 0.01) with 40◦ knee flexion. abducted/valgus and in a relatively externally rotated position, which is in line with previous findings (Yoneyama and Okamoto, Knee Joint Kinetics 2000). The increase of the external rotation with an increase of In all cases, the alignment of GRF in the frontal plane was medial the waist width was probably the result of the active adaptation of the knee joint center, with no significant difference between of the skier to reduce the external knee joint torque in the frontal ◦ different waist widths. With 40 knee flexion, the distances (r) plane. The increase of external rotation was in accordance with between the knee joint center and GRF were 1.58 ± 1.96 (W0), the previous field study (Zorko et al., 2015). On the other hand, 1.85 ± 2.13 (W60), 1.38 ± 2.08 (W90), and 1.06 ± 1.86 cm knee abduction remained the same with any waist width. One ◦ (W120). With 60 knee flexion, the distances were 1.38 ± 2.01 might also expect an increment of the abduction with a greater (W0), 1.48 ± 2.48 (W60), 0.84 ± 2.15 (W90), and 1.42 ± 2.54 cm waist width. Greater abduction could possibly bring the knee (W120). The magnitude of GRF throughout the experiment closer to the GRF vector, that is, avoiding an increase of varus remained at the same value as was set at the start (110% torque. One possible explanation for relatively fixed abduction body weight, Figure 5). Similarly stands for the angle of GRF with any waist width could be that the knee reached the near (Figure 6). As GRF ran medially to the knee joint center, torque to end range of motion even with the narrowest skis, which ◦ (M) acted on the knee in the varus direction. The distribution is reported to be 5–10 in the frontal plane for a loose joint of M for different waist widths is presented in Figure 7. There position (Grood et al., 1988), and thus, any further increments were no statistical significant differences in magnitude of M with of abduction were impossible. Instead, a combination of flexion different waist widths for either knee joint flexion settings. and external rotation in the knee joint with a fixed longitudinal

Frontiers in Sports and Active Living | www.frontiersin.org 5 February 2020 | Volume 2 | Article 7 Zorko et al. Wide Skis and Knee Injury

FIGURE 7 | Knee joint torques in the frontal plane (mean ± SD) with knee flexion 40◦ (left diagram) and 60◦ (right diagram). There were no statistical differences in knee torques between different ski waist widths.

TABLE 1 | The RMS values of the surface EMG attained during different ski widths (normalized to maximal RMS attained during MVC).

W0 W60 W90 W120

40◦ knee flexion BF 12 ± 6% 17 ± 10%** 24 ± 16%**,## 35 ± 25%**,## ST 8 ± 4% 6 ± 4% 7 ± 4% 8 ± 6% 60◦ knee flexion BF 16 ± 9% 17 ± 9%** 19 ± 10%* 24 ± 16%**,## ST 15 ± 8% 12 ± 8%** 11 ± 6%** 11 ± 7%**

Asterisks indicate significant differences from central position values (**P < 0.01; *P < 0.05). Number signs indicate significant differences from narrower position values (##P < 0.01). BF, biceps femoris; ST, semitendinosus; RMS, root mean square; EMG, electromyography; MVC, maximal voluntary contraction.

distance from the GRF vector to the center of the knee joint) also remained the same, despite the point of application of GRF between the narrowest (60 mm) and widest (120 mm) ski width having a displacement of 3 cm. This was only possible with the previously described knee kinematic changes. GRF “pierced” the medial knee joint compartment in the frontal plane within very tight values, 1–2cm medially to the joint center, which is in accordance with the physiological alignment of GRF by single leg stance (Levine and Bosco, 2007) as well as in accordance with general human limb malalignment adaptive strategies (Mündermann et al., 2005; Levine and Bosco, 2007). FIGURE 8 | The activation of biceps femoris (BF) and semitendinosus (ST) Obviously, in possible adverse biomechanical conditions (such (mean ± SD) with 40◦ (A) and 60◦ (B) knee flexion. *Depicts statistical significant difference against W(0), and #depicts statistical significant as using very wide skis on a hard snow base during turns), the difference with the previous (narrower) waist width. body attempts to retain the unchanged alignment of GRF in relation to knee compartments. This is done by adapting the knee kinematics. Perhaps this is also an attempt to limit the change of orientation of the ski resulted in the medial movement of the the external torque acting on the knee joint. knee joint. Electromyographic Findings Knee Joint Kinetics The increment of activation of BF with the waist width increment Torque of the outside knee joint was not influenced by the waist is in accordance with the changes in the knee external rotation as width of the ski. The angle and the magnitude of GRF in the biceps is also an external rotator of the knee (Besier et al., 2003). ◦ ◦ frontal plane remained the same throughout the experiment. The The bigger activation at 40 knee flexion compared with 60 , knee joint moment arm in the frontal plane (the orthogonal despite a smaller achieved external rotation, is perhaps due to

Frontiers in Sports and Active Living | www.frontiersin.org 6 February 2020 | Volume 2 | Article 7 Zorko et al. Wide Skis and Knee Injury

◦ a less feasible rotational knee joint movement at 40 of flexion, the femur when using wider waist width skis. However, the where the ligaments are in a tighter position (Platzer, 2004). valgus position of the knee remained independent of the ski The magnitude and the ratio of the outer and inner hamstring waist width. The intention of these kinematical adjustments muscle activations in our study were comparable with those was probably to maintain alignment of the GRF of the loaded of the average activation of these muscles in a previous field lower limb within close proximity to the joint center, thus study (Nemeth et al., 1997). Pronounced oscillatory patterns of minimizing the external torques acting on the joint. Such lower limb muscle activation have been reported in real skiing minimization of the external torques while using very wide waist situations (Panizzolo et al., 2010), so only average values could be skis resulted in the knee joint reaching its near end of the compared in the present simulation. range of motion in the frontal and transversal planes combined with higher muscle activation. The probable consequence of Methodological Considerations using skis with a very large waist width on hard frozen surfaces The crucial concern is whether the laboratory settings sufficiently would be that the knee joint is continuously (during numerous simulated real alpine skiing turns. The obvious mechanical turning) in an externally rotated position and femoral muscles difference is that the participants had no movement along the becoming more activated with possible more compression forces sagittal axis. However, taking into account that biomechanical acting on joint surfaces. Previous studies (Mündermann et al., parameters were only in the frontal plane, where at any specific 2005; Levine and Bosco, 2007) clarified that the knee joint moment in time there is also no movement in real skiing, malalignment is a well-known risk factor for degenerative knee the performed simulation seems to be suitable. Furthermore, joint conditions. However, whether this type of malalignment the simulation design allowed some field perturbations to be and additional muscle activation can lead to long-term knee excluded, as well as provided more superior standardization of joint consequences in skiing is yet unclear. Nevertheless, very the measurement conditions compared with those in a previous large waist widths are not permitted in most competitive study dealing with ski width (Zorko et al., 2015): (i) identical alpine skiing disciplines, although International Ski Association and continuously maintained ski and center of mass inclination regulations still have no upper waist width limit in slalom, angle ensured simulation of the same virtual turning radius for which has already been pointed out as a potential risk factor each participant, (ii) by keeping a stationary knee flexion angle, (Supej et al., 2017). the influence of magnitude of flexion on rotation and abduction angle (Wilson et al., 2000; Lu et al., 2008) and the influence on the Perspectives lower limb muscle activations were excluded. This is the first study to investigate knee joint kinematics, At first glance, it may seem that the limitation of the study kinetics, and muscle activation in alpine skiing as a function was also the setting of the skier in a manner that their center of the ski waist width. Rotation of the tibia against the femur of mass did not move during the simulation for different was shown to be progressively influenced by the increment widths of the skis. In other words, the strap giving lateral of the waist width of the outer ski during turning. Therefore, support to the participant remained unchanged during the sets increasing waist width potentially escalates uneven joint pressure of movements. One might conclude that all the kinematic and distribution, while at the same time, this adaptation allows the torque changes were merely the consequence of these setting external torque to remain unchanged. Future epidemiological changes. However, it was proven that such small changes of the studies are needed to further elucidate the potential relationship moving foot relative to the fixed point of the side attachment between ski waist width and the damaging effect on the had a negligible effect on the magnitude (Figure 5) as well as on knee joint. Additionally, further improvements in the ski the alignment of the GRF (Figure 6). Consequently, this setting simulations are needed, possibly integrating dynamic movements could not have any influence on the torques of the individual for imitating turning as well as vibrations, to better simulate real body segments. skiing conditions. Perhaps the single leg stance in our experiment may represent an important difference compared with that in real skiing, where typically both legs are loaded. However, it was shown that the DATA AVAILABILITY STATEMENT predominant load is normally shifted to the outer ski (Vaverka Publicly available datasets were analyzed in this study. This and Vodickova, 2010). This still might have an impact on the data can be found here: https://plus.si.cobiss.net/opac7/bib/ knee torque situation (Klous et al., 2012), which might be 288145920. different in real skiing. Nevertheless, the loading of the outer leg in our experiment was comparable with the expected average loading of the outer leg during the turn in recreational skiing, if ETHICS STATEMENT we assume 2:1 force distribution between the outer and inner legs (Vaverka and Vodickova, 2010; Scheiber et al., 2012). This study was carried out in accordance with the recommendations of the responsible Ethics Committee at Summary the University of Ljubljana with written informed consent The present study has demonstrated that skiers adapt knee joint from all subjects. All subjects gave written informed consent in kinematics with additional external rotation of the tibia against accordance with the Declaration of Helsinki. The protocol was

Frontiers in Sports and Active Living | www.frontiersin.org 7 February 2020 | Volume 2 | Article 7 Zorko et al. Wide Skis and Knee Injury approved by the responsible Ethics Committee at the University FUNDING of Ljubljana. This study was supported financially by the Foundation for AUTHOR CONTRIBUTIONS Financing Sport Organisations in Slovenia (RR-14-004, RR-13- 145, RR-12-036, and RR-11-108), University of Ljubljana (P2- MS and MZ designed the study, except for the neurophysiological 0228) and the Slovenian Research Agency. part, which was designed by KT. MZ, MS, and KT conducted the experiment and collected all the data. BN developed the software for 3D kinematical analysis. ZM and AO developed the skiing ACKNOWLEDGMENTS simulator and supervised the experimental process. All authors have contributed in the writing of the manuscript, proofread the The authors would like to sincerely thank the subjects for manuscript, and approved the final version. their participation.

REFERENCES Levine, H. B., and Bosco, J. A. III. (2007). Sagittal and coronal biomechanics of the knee: a rationale for corrective measures. Bull. NYU Hosp. Jt. Dis. 65, 87–95. Bere, T., Florenes, T. W., Krosshaug, T., Koga, H., Nordsletten, L., Irving, C., et al. Lu, T. W., Tsai, T. Y., Kuo, M. Y., Hsu, H. C., and Chen, H. L. (2008). In vivo three- (2011). Mechanisms of anterior cruciate ligament injury in World Cup alpine dimensional kinematics of the normal knee during active extension under skiing: a systematic video analysis of 20 cases. Am. J. Sports Med. 39, 1421–1429. unloaded and loaded conditions using single-plane fluoroscopy. Med. Eng. doi: 10.1177/0363546511405147 Phys. 30, 1004–1012. doi: 10.1016/j.medengphy.2008.03.001 Besier, T. F., Lloyd, D. G., and Ackland, T. R. (2003). Muscle activation strategies Mündermann, A., Dyrby, C. O., and Andriacchi, T. P. (2005). Secondary gait at the knee during running and cutting maneuvers. Med. Sci. Sports Exerc. 35, changes in patients with medial compartment knee osteoarthritis: increased 119–127. doi: 10.1097/00005768-200301000-00019 load at the ankle, knee, and hip during walking. Arthritis Rheum. 52, Brucker, P. U., Katzmaier, P., Olvermann, M., Huber, A., Waibel, K., Imhoff, 2835–2844. doi: 10.1002/art.21262 A. B., et al. (2014). Recreational and competitive alpine skiing. Typical Nemeth, G., Lamontagne, M., Tho, K. S., and Eriksson, E. (1997). injury patterns and possibilities for prevention. Der Unfallchirurg 117, 24–32. Electromyographic activity in expert downhill skiers using functional doi: 10.1007/s00113-013-2464-4 knee braces after anterior cruciate ligament injuries. Am. J. Sports Med. 25, Federolf, P., Roos, M., Lüthi, A., and Dual, J. (2010). Finite element simulation 635–641. doi: 10.1177/036354659702500508 of the ski–snow interaction of an alpine ski in a . Sports Eng. 12, Panizzolo, F. A., Petrone, N., and Marcolin, G. (2010). Comparative analysis of 123–133. doi: 10.1007/s12283-010-0038-z muscle activation patterns between skiing on slopes and on training devices. Federolf, P., Tscharner, V., Häufle, D., Nigg, B., Gimpl, M., and Müller, E. (2009). Proc. Eng. 2, 2537–2542. doi: 10.1016/j.proeng.2010.04.028 “Vibration exposure in alpine skiing and consequences for muscle activation Platzer, W. (2004). Color Atlas of Human Anatomy: Locomotor System, 5th Edn. levels,” in Science & Skiing, IV, eds E. Müller, S. Lindinger, and T. Stöggl Stuttgart: Thieme. (Maidenhead: Meyer & Meyer Sport (UK) Ltd.), 19–25. Scheiber, P., Seifert, J., and Müller, E. (2012). Relationships between biomechanics Gilgien, M., Sporri, J., Kroll, J., Crivelli, P., and Muller, E. (2014). Mechanics of and physiology in older, recreational alpine skiers. Scand. J. Med. Sci. Sports 22, turning and jumping and skier speed are associated with injury risk in men’s 49–57. doi: 10.1111/j.1600-0838.2010.01146.x World Cup alpine skiing: a comparison between the competition disciplines. Sharma, L., Song, J., Felson, D. T., Cahue, S., Shamiyeh, E., and Dunlop, D. Br. J. Sports Med. 48, 742–747. doi: 10.1136/bjsports-2013-092994 D. (2001). The role of knee alignment in disease progression and functional Grood, E. S., Stowers, S. F., and Noyes, F. R. (1988). Limits of movement decline in knee osteoarthritis. JAMA 286, 188–195. doi: 10.1001/jama.286.2.188 in the human knee. Effect of sectioning the posterior cruciate ligament Shea, K. G., Archibald-Seiffer, N., Murdock, E., Grimm, N. L., Jacobs, J. C., Willick, and posterolateral structures. J. Bone Joint Surg. Am. 70, 88–97. S., et al. (2014). Knee injuries in downhill skiers: a 6-year survey study. Orthop doi: 10.2106/00004623-198870010-00014 J. Sports Med. 2:2325967113519741. doi: 10.1177/2325967113519741 Grood, E. S., and Suntay, W. J. (1983). A joint coordinate system for the clinical Sporri, J., Kroll, J., Gilgien, M., and Muller, E. (2016). Sidecut radius and the description of three-dimensional motions: application to the knee. J. Biomech. mechanics of turning-equipment designed to reduce risk of severe traumatic Eng. 105, 136–144. doi: 10.1115/1.3138397 knee injuries in alpine ski racing. Br. J. Sports Med. 50, 14–19. Haaland, B., Steenstrup, S. E., Bere, T., Bahr, R., and Nordsletten, L. (2016). doi: 10.1136/bjsports-2015-095737 Injury rate and injury patterns in FIS World Cup Alpine skiing (2006-2015): Sporri, J., Kroll, J., Haid, C., Fasel, B., and Muller, E. (2015). have the new ski regulations made an impact? Br. J. Sports Med. 50, 32–36. Potential mechanisms leading to overuse injuries of the doi: 10.1136/bjsports-2015-095467 back in alpine ski racing: a descriptive biomechanical study. Hebert-Losier, K., and Holmberg, H. C. (2013). What are the exercise-based injury Am J. Sports Med. 43, 2042–2048. doi: 10.1177/03635465155 prevention recommendations for recreational alpine skiing and snowboarding? 88178 A systematic review. Sports Med. 43, 355–366. doi: 10.1007/s40279-013-0032-2 Supej, M. (2013). “Vibrations in recreational alpine skiing: a pilot study,” in Hermens, H. J., Freriks, B., Disselhorst-Klug, C., and Rau, G. (2000). Development ISBS-Conference Proceedings Archive 1. 31th International Conference on of recommendations for SEMG sensors and sensor placement procedures. J. Biomechanics in Sports (Taipei). Electromyogr. Kinesiol. 10, 361–374. doi: 10.1016/S1050-6411(00)00027-4 Supej, M., Kugovnik, O., and Nemec, B. (2004). Modelling and simulation of two Hildebrandt, C., and Raschner, C. (2013). Traumatic and overuse injuries among competition slalom techniques. Kinesiology 36, 206–212. elite adolescent alpine skiers: a two-year retrospective analysis. Int. Sport Med. Supej, M., Senner, V., Petrone, N., and Holmberg, H. C. (2017). Reducing the risks J. 14, 245–255. for traumatic and overuse injury among competitive alpine skiers. Br. J. Sports Klous, M., Müller, E., and Schwameder, H. (2010). Collecting kinematic data on Med. 51, 1–2. doi: 10.1136/bjsports-2016-096502 a ski/ track with panning, tilting, and zooming cameras: is there Urabe, Y., Ochi, M., Onari, K., and Ikuta, Y. (2002). Anterior cruciate sufficient accuracy for a biomechanical analysis? J. Sports Sci. 28, 1345–1353. ligament injury in recreational alpine skiers: analysis of mechanisms and doi: 10.1080/02640414.2010.507253 strategy for prevention. J. Orthop. Sci. 7, 1–5. doi: 10.1007/s776-002- Klous, M., Müller, E., and Schwameder, H. (2012). Three-dimensional knee joint 8405-7 loading in alpine skiing: a comparison between a carved and a skidded turn. J. Vaverka, F., and Vodickova, S. (2010). Laterality of the lower limbs and carving Appl. Biomech. 28, 655–664. doi: 10.1123/jab.28.6.655 turn. BiolSport 27, 129–134. doi: 10.5604/20831862.913080

Frontiers in Sports and Active Living | www.frontiersin.org 8 February 2020 | Volume 2 | Article 7 Zorko et al. Wide Skis and Knee Injury

Wilson, D. R., Feikes, J. D., Zavatsky, A. B., and O’Connor, J. J. Conflict of Interest: The authors declare that the research was conducted in the (2000). The components of passive knee movement are coupled to absence of any commercial or financial relationships that could be construed as a flexion angle. J. Biomech. 33, 465–473. doi: 10.1016/S0021-9290(99)0 potential conflict of interest. 0206-7 Yoneyama, K., and Okamoto, S. (2000). Joint motion and reacting Copyright © 2020 Zorko, Nemec, Matjaˇci´c, Olenšek, Tomazin and Supej. This is an forces in the carving ski turn compared with the conventional open-access article distributed under the terms of the Creative Commons Attribution ski turn. Sports Eng. 3, 161–176. doi: 10.1046/j.1460-2687.2000.0 License (CC BY). The use, distribution or reproduction in other forums is permitted, 0060.x provided the original author(s) and the copyright owner(s) are credited and that the Zorko, M., Nemec, B., Babic, J., Lesnik, B., and Supej, M. (2015). The waist width original publication in this journal is cited, in accordance with accepted academic of skis influences the kinematics of the knee joint in alpine skiing. J. Sports Sci. practice. No use, distribution or reproduction is permitted which does not comply Med. 14, 606–619. with these terms.

Frontiers in Sports and Active Living | www.frontiersin.org 9 February 2020 | Volume 2 | Article 7